Bump hunting in latent space
Blaž Bortolato, Aleks Smolkovič, Barry M. Dillon, Jernej F. Kamenik
Abstract
Unsupervised anomaly-detection could be crucial in future analyses searching for rare phenomena in large datasets, as for example collected at the LHC. To this end, we introduce a physics inspired variational autoencoder (VAE) architecture which performs competitively and robustly on the LHC Olympics Machine Learning Challenge datasets. We demonstrate how embedding some physical observables directly into the VAE latent space, while at the same time keeping the anomaly-detection manifestly agnostic to them, can help to identify and characterize features in measured spectra as caused by the presence of anomalies in a dataset.
Topics & Concepts
AutoencoderEmbeddingAnomaly detectionLarge Hadron ColliderObservableArtificial intelligenceClassifier (UML)Computer scienceUnsupervised learningSpace (punctuation)Machine learningArchitectureDeep learningPattern recognition (psychology)PhysicsGeographyParticle physicsOperating systemArchaeologyQuantum mechanicsComputational Physics and Python ApplicationsParticle physics theoretical and experimental studiesAnomaly Detection Techniques and Applications